1,2 2 3 2 ACCEPTED -...
Transcript of 1,2 2 3 2 ACCEPTED -...
Journal of Strength and Conditioning Research Publish Ahead of PrintDOI: 10.1519/JSC.0000000000001968
Title of the Article: Effect of Body Position on Force Production During the Isometric Mid-1 Thigh Pull 2 Submission Type: Original Investigation 3
4
George K. Beckham1,2, Kimitake Sato2, Satoshi Mizuguchi2, G. Gregory. Haff3, Michael H. 5 Stone2 6 1: Kinesiology Department, California State University, Monterey Bay, Seaside, CA. 7 2: Center of Excellence for Sport Science and Coach Education, Department of Exercise and 8 Sport Science, East Tennessee State University, Johnson City, TN. 9 3: Center for Exercise and Sport Science Research, Edith Cowan University, Joondalup, 10 Australia. 11
12
Corresponding Author 13
George K. Beckham 14 [email protected] 15 Phone: (831)582-5258 16 Fax: (831) 582-3737 17 100 Campus Center 18 Valley Hall D 19 Seaside, CA 93955 20
Preferred Running Head: Isometric Mid-Thigh Pull Body Position 21
Author Statement: 22
The following manuscript has been read and approved by all of the listed co-authors and meets 23 the guidelines of co-authorship. No funding was received for this study. 24
Conflicts of Interest: 25
The authors have no conflicts of interest to report. 26
Abstract word count: 199 27
Number of Figures and Tables: 7 28
29
30
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
ABSTRACT 32
33
Various body positions have been used in the scientific literature when performing the 34
isometric mid-thigh pull resulting in divergent results. We evaluated force production in the 35
isometric mid-thigh pull in bent (125° knee and 125° hip angles) and upright (125° knee, 145° 36
hip angle) positions in subjects with (>6 months) and without (< 6 months) substantial 37
experience using weightlifting derivatives. A mixed-design ANOVA was used to evaluate the 38
effect of pull position and weightlifting experience on peak force, force at 50ms, 90ms, 200ms, 39
and 250ms. There were statistically significant main effects for weightlifting experience and pull 40
position for all variables tested, and statistically significant interaction effects for peak force, 41
allometricaly scaled peak force, force at 200ms, and force at 250ms. Calculated effect sizes were 42
small to large for all variables in subjects with weightlifting experience, and were small to 43
moderate between positions for all variables in subjects without weightlifting experience. A 44
central finding of the study is that the upright body position (125° knee and 145° hip) should be 45
used given that forces generated are highest in that position. Actual joint angles during maximum 46
effort pulling should be measured to ensure body position is close to the position intended. 47
48
Key words: maximum strength, performance testing, weightlifting, test variability, strength 49
testing 50
51
52
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
INTRODUCTION 53
Maximal strength testing is a valuable method for evaluating athletes and monitoring 54
training adaptations (19). While maximal strength is commonly assessed using 1-repetition 55
maximum testing (1-RM), other means of evaluating maximal strength have recently been 56
suggested to be equally or more efficacious and efficient (19). One such method is the isometric 57
mid-thigh pull (IMTP), which has significant advantages over 1-RM testing, such as less time 58
spent testing, reduced training volume, reduced accumulated fatigue, and potentially being safer 59
than 1-RM testing. Additionally, strong correlations between IMTP variables and dynamic 60
movements such as the 1-RM back squat, snatch and clean have been reported (2, 21). 61
The original research on the IMTP selected a body position that mimicked the second 62
pull of the clean (9). This position was selected because the highest forces and bar velocities are 63
generated during this phase of weightlifting movements (14). However, as the IMTP has 64
increased in popularity as a performance test, there have been inconsistencies in the methods 65
used for performing the test. In particular, the precise posture and body position used in previous 66
studies has not always accurately represented the second pull position used in the original 67
research. Most studies have used a knee angle of approximately 120°-135° (2, 3, 8, 12, 15, 21), 68
while there has been substantial variability in what hip angle has been used in the studies that 69
have reported it (several studies have not reported the hip angle used in testing) (16-18). 70
To examine the question of whether the hip and knee angle used during the IMTP affects 71
force production, Comfort et al. (6) evaluated changes in force production between 9 different 72
hip and knee angles, and found that there were no differences between each position. Another 73
study by Beckham et al. (3) found conflicting results in powerlifters, who had higher peak forces 74
in an upright torso IMTP position compared to a bent-over torso position (neutral spine, but 75
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
greater hip flexion). Given that these two studies offer divergent findings, there is not consensus 76
in the scientific literature on the impact of body position on forces generated during the IMTP. 77
If body position does influence the force produced during the test, then it becomes 78
substantially more difficult to compare the results between studies that use different body 79
positions. Additionally, within the same study, if there is a large variation between subjects in the 80
body positions used, then an additional source of measurement error and variability in force 81
production has been included in IMTP measurement making it more difficult to draw accurate 82
conclusions about a group of subjects’ performance capacities. However, if body position does 83
not result in differences in force production, then it is not a source of error in the prior two 84
scenarios. 85
Experience with weightlifting derivatives (derivatives of the snatch and clean, e.g. mid-86
thigh pulls, power cleans or power snatches)(22) may exert a potential influence on the impact of 87
body position on force production during the IMTP test. Generally, weightlifters produce the 88
highest forces during the second pull phase of the clean and perform training exercises to 89
maximize their abilities in this position (10). Because the IMTP was originally based on a 90
position similar to the second pull of the clean, it is plausible that significant training experience 91
with the second pull position would influence how effective one might be with that pulling 92
position. Thus, it is worthwhile to evaluate the potential influence of experience with 93
weightlifting derivatives on force production differences between body positions. 94
The purpose of this study was to evaluate the effects of body position and experience 95
with weightlifting movements on force production in isometric mid-thigh pull. We hypothesized 96
that the upright body position would result in higher forces generated during the IMTP, and that 97
experience with weightlifting derivatives would increase this difference. 98
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
99
METHODS 100
EXPERIMENTAL APPROACH TO THE PROBLEM 101
The present study was conducted in two parts in order to examine the impact of body 102
position and training experience on the performance of the IMTP. For part 1 of this study, the 103
use of an upright and bent body position during the IMTP were evaluated in subjects with and 104
without experience with weightlifting or weightlifting derivatives. Part 2 of the study was a 105
second data collection and data analysis period that was performed while attempting to use the 106
IMTP methods specifically outlined by Comfort et al.(6). This was undertaken in an attempt to 107
replicate Comfort et al.’s (6) findings and compare to the findings of part 1 of the present study. 108
109
PART 1: SUBJECTS 110
Two groups of subjects were recruited for this study. All subjects, regardless of group 111
were required to be male and involved in regular physical activity. One group had greater than 6 112
months of experience training with weightlifting movements. This group was designated the 113
“experience with weightlifting” group (n=12, body mass: 84.4±7.4kg, years of weightlifting: 114
4.9±4.2y range: 1.07-13.5y). The other group, with less than 6 months experience training with 115
weightlifting movements, was designated the “low experience with weightlifting” group (n=10, 116
body weight: 75.1±11.5kg, years of weightlifting: 0.09±0.09y range: 0.00-0.24y). Prior to 117
participation, all subjects were thoroughly informed of study procedures. Each subject then read 118
and signed informed consent documents according to procedures outlined by the University 119
Institutional Review Board and in accordance with the Declaration of Helsinki. All subjects were 120
free from musculoskeletal injury for at least 6 months prior to testing. 121
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
PART 1: PROCEDURES 122
Subjects came into the laboratory on 5 separate occasions, separated by 72-96 hours. In 123
each session, subjects performed the IMTP in a custom-designed power rack (Sorinex, Irmo, SC) 124
that allows the bar to be fixed at any height, while standing on two adjacent force plates (45.5 cm 125
x 91 cm, RoughDeck HP; Rice Lake Weighing Systems). Subjects were secured to the bar using 126
lifting straps and athletic tape in accordance with previous methods.(9) 127
In the first testing session, bar heights and foot position were determined and recorded for 128
the upright and bent positions so that they could be replicated in each subsequent session. The 129
bar heights to allow for each body position were determined in the first testing session by using a 130
digital camera (HD Pro Webcam C920, Logitech Inc.) and freely available angle measurement 131
software (Screen Scales, Talon Designs LLP). When initially measuring bar heights and joint 132
angles in the first familiarization session, subjects were instructed to pull on the bar with 50% 133
effort in an effort to remove slack from the body. 134
Subjects performed IMTP trials in each session as outlined by Figure 1. Procedures were 135
identical on each day of testing, except that the pull position order was randomized to remove 136
testing order bias. Only data collected on the fifth and final session were used for this study. 137
Two separate pulling positions were evaluated during the IMTP in randomized order. 138
Specifically, a body position which allowed a knee angle of 125° and hip angle of 145° was 139
designated the “upright” position, and a body position which allowed a knee angle of 125° and 140
hip angle of 125° was designated the “bent” position. The knee angle of 125° represents the 141
approximate angle commonly used in IMTP studies (2, 3, 8, 12, 15, 21). The two hip angles were 142
meant to approximate the upright body position used in many studies (2, 15, 21), while the bent 143
position was meant to approximate the body position used in others (6, 16). 144
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
On each testing day, subjects performed a standardized warmup of 2 minutes of cycling 145
at 50 watts with 50 to 60 RPM. Subjects then performed 6 repetitions each of: forward walking 146
lunges, reverse walking lunges, side lunges, straight leg march, and quadriceps pulls, then 5 147
bodyweight squats and 5 ballistic bodyweight squats. This standard warmup was specifically 148
chosen to reduce the possibility that the warmup would preferentially benefit either pulling 149
position. After the warmup, the order, intensity and rest of IMTPs went according to procedures 150
outlined in Figure 1. 151
152
Figure 1 about here 153
154
To ensure there was minimal slack in the body before initiation of the pull, subjects were 155
instructed to use a minimal of pre-tension (2). Once in position (verified by viewing the subject 156
and stability of the force trace), subjects received a countdown to begin the pull and were 157
instructed when to stop in accordance with previous methods (9). For all maximum effort pulls, 158
subjects received substantial encouragement by the investigators to ensure a maximal effort. 159
Before each pull, subjects were instructed to “pull as hard and fast as possible” to maximize rate 160
of force development and peak force (4). 161
On sessions 1, 3, and 4, subjects only performed two maximal effort pulls, while on 162
sessions 2 and 5, subjects performed between 2 and 4 pulls. Ideally, subjects needed only to 163
perform 2 pulls on sessions 2 and 5, but maximum effort attempts were repeated if errors in 164
pulling were observed (countermovement or a substantial change in body position) or if a ≥250N 165
difference in peak force were measured (9, 15). If 4 trials were needed, the best 2 trials were 166
used for analysis. Only the data from testing session 5 was used for the present study. 167
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Analog data from the force plate were amplified and low-pass filtered at 16 Hz 168
(Transducer Techniques, Temecula, CA), and sampled at 1000 Hz (DAQCard-6063E, National 169
Instruments). Force-time curves were digitally filtered using a 2nd order Butterworth low-pass 170
filter at 10 Hz and analyzed using a custom Labview program (Labview 2010, National 171
Instruments). The following variables were calculated from the force time curve generated 172
during each pull, peak force (PF), and force at various time points after the initiation of the pull 173
including force at 50ms (F50), force at 90ms (F90), force at 200ms (F200), and force at 250ms 174
(F250). The start of each pull was identified by visual inspection. In addition, peak force was 175
scaled allometrically to account for body mass differences, using the equation force·bodymass-176
0.67 (13). 177
Sagittal plane video was recorded for each pull (HD Pro Webcam C920, Logitech Inc.). 178
Joint angles for the knee and hip were evaluated at the start (just before initiation of the pull), 179
and most extreme (point at which joint angles were at their maximum during the pull). 180
181
PART 1: STATISTICAL ANALYSIS 182
All data were screened for within-session test-retest reliability, outliers and normality. 183
Reliability was assessed using ICCs with 95% CI, and CV with 95% CI (typical error of log-184
transformed data). Each reliability metric was calculated on the entire group, as well as each 185
subset of data (group and position). Data were also screened for violations of assumptions for a 186
mixed-design ANOVA (23). 187
Multiple 2x2 mixed ANOVAs (weightlifting experience X pulling position) were run to 188
determine differences between groups and position for each variable tested. Generalized eta-189
squared (ηg2) was used for effect sizes and interpreted with the following scale: 0.02 small, 0.13 190
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
medium, and 0.26 large (1, 5). In Study 1, The Hedge’s g correction for small sample sizes of 191
Cohen’s d effect size statistics were calculated between pulling positions for the experienced and 192
inexperienced groups. The magnitude of effect sizes was interpreted according to a scale by 193
Hopkins (11) as follows: 0 trivial, 0.2 small, 0.6 moderate, 1.2 large, and >2.0 very large. All 194
analysis was performed in R (The R Foundation for Statistical Computing, Vienna Austria), 195
using the ‘psych’, ‘effsize’, ‘pastecs’ and ‘ezANOVA’ analysis packages (20). 196
197
PART 1: RESULTS 198
PF, F50, F90, F200, and F250 were adequately reliable for later analysis. Reliability 199
statistics can be found in Table 1. Descriptive statistics for IMTP variables can be found in Table 200
2. 201
202
Insert Table 1 and Table 2 About Here 203
204
Specific results from each of the repeated measures ANOVAs can be found in Table 3. 205
Pairwise effect size statistics between pulling positions are found in Figure 2. The statistical 206
interaction effect showed the general pattern that experienced lifters produced greater values in 207
the upright position than inexperienced lifters in PF, PFa, F200, and F250. For the variables F50 208
and F90, while an interaction effect was not present, there was a main effect for position, 209
indicating greater values in the upright position when both experience groups were combined. 210
Furthermore, regardless of the presence of a statistical interaction effect, all variables showed a 211
moderate or small effect in favor of the upright position in the experienced and the inexperienced 212
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
groups, respectively, although the 95% confidence intervals of the two groups overlapped 213
(Figure 2). 214
Sagittal plane angle data for the hip and knee for each IMTP position are reported in 215
Table 4. Small amounts of extension during the pull were observed for the knee and hip for both 216
the bent and upright pulling positions. 217
218
Insert Table 3 and Table 4 About Here 219
220
Insert Figure 2 About Here 221
222
223
PART 2: METHODS 224
PART 2: SUBJECTS 225
Subjects for Study 2 were experienced with both weightlifting (>6 months) and the IMTP 226
in both positions. Subjects were fully informed about study procedures, and gave their informed 227
consent to participate. A total of 8 subjects were initially recruited for testing, however two 228
subjects were unable to achieve positions outlined above. Specifically, these two subjects were 229
unable to achieve the prescribed position and maintain the bar position in alignment with the 230
thigh mark. Another subject increased his hip angle from 125° to 140° during the bent pull, and 231
was therefore excluded on the basis that this did not represent the bent position. Thus, force data 232
for five subjects were analyzed. 233
234
235
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Part 2 of this study was performed after the completion of part 1 of this study when it was
236
determined that part 1 had different findings than those reported by Comfort et al. (6) on the
237
impact of knee and hip angle on IMTP force-time curve variables. Statistically significant
238
differences between testing positions for all variables tested were observed in part 1, but
239
differences were not found in the study by Comfort et al. (6). Slight changes in positioning and
240
set up were made after further comparison between the methods of part 1 and correspondence
241
with the Comfort et al. (6) in order to ensure a more accurate replication of their original work.
242
To evaluate if differences in findings between part 1 and the study by Comfort et al. (6) were due
243
to slight differences in bar positioning on the thigh between both studies (despite similar knee
244
and hip angles used in both studies), the following changes to testing procedures were introduced
245
for part 2 based upon direct communication with Comfort et al. (6) about their research:
246
1. A horizontal line was drawn in marker across the thighs marking exactly half the distance
247
between the anterior superior iliac spine and center of the patella. When setting up the
248
subject within the custom power rack, the bar had to cover the line drawn on the thigh.
249
2. Foot movement was not allowed to deviate between the two body positions.
250
All IMTPs were performed in a single session, with each pull position performed in
251
randomized order. Subjects’ thighs were marked as outlined above and each entered the rack to
252
measure bar heights for each position. Bar heights and joint angles were determined in the same
253
manner as to Part 1. Warmups, rest periods and maximum effort pulls were structured identically
254
to methods used in Part 1.
255
256
257
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
PART 2: PROCEDURES
Absolute differences were calculated between each position on an individual basis so that
258
individual changes between positions could be quantified. Hedge’s g and 95% confidence
259
interval was calculated for the group changes.
260
261
PART 2: RESULTS
262
Comparisons of results between each pulling position can be found in Table 5. Each
263
subject improved performance in the upright position for nearly all variables measured, with
264
three subjects improving in all variables measured. Effect size and 95% confidence interval
265
between positions (negative effect size indicating that values for the upright position were larger)
266
for PF, F50, F90, F200, and F250 were: -0.59 (-1.86-0.67), -0.19 (-1.43-1.05), -0.35 (-1.06-0.9), -
267
0.54 (-1.81-0.72), and -0.63 (-1.9-0.64).
268
269
Table 5 & 6 about here 270
271
DISCUSSION 272
The main findings of this two-part study are that there are differences in the force 273
production capabilities for subjects when performing the IMTP with different body positions. 274
More specifically, the upright position appears to be the position in which subjects can create 275
higher forces more quickly. The magnitude of force production difference between the bent and 276
upright positions does depend on whether subjects are experienced with weightlifting or not, as 277
indicated by the statistically significant interaction effect, and the generally lower effect sizes 278
between pulling positions for the subjects with less experience with weightlifting derivatives. 279
Subjects who are experienced with weightlifting exhibit greater differences between the two 280
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
PART 2: STATISTICAL ANALYSIS
positions, as indicated by the moderate to large effect sizes observed (g = 0.6-1.2). Subjects 281
without weightlifting experience still exhibited differences in force generation capacity between 282
the two positions as indicated by the small to moderate effect sizes (g = 0.3-0.8) between 283
positions. 284
From a specificity perspective, it is understandable that the weightlifting-experienced 285
group would perform better in the body position that mimics the second pull of the clean or 286
snatch (upright position). The phase of the clean and snatch with the highest forces is the second 287
pull (10), which is represented by the upright position used in the present study and previously 288
published research (8). Since weightlifters frequently train with exercises that require and 289
develop mastery of this position it is possible that they have maximized their ability to develop 290
forces in this position. It is not unexpected that the bent over position results in reduced force 291
production as it corresponds to the transition phase which links the first and second pull in 292
weightlifting movements. Overall, the transition phase of the pulling motion always exhibits the 293
lowest forces as a result of the mechanical disadvantages associated with the position in 294
weightlifting (10). Conceptually, the transition phase functions to reposition the body and 295
prepare the lifter for execution of the second pull where she or he is able to maximize force 296
generation (7). The increased force production may be due to better mechanical advantage, 297
muscle lengths, and potentially engagement of the stretch-shortening cycle, although only the 298
former two factors would be afforded to force production in the IMTP, given its isometric 299
execution. 300
For subjects with less weightlifting experience, it would make sense that there is a 301
reduced difference between the tested positions. These subjects would have spent less time (if 302
any time at all) overloading the power position and second pull, and would not be expected to 303
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
display the effects of training this position. There is however, still an apparent mechanical 304
advantage when using the upright position even among those subjects who are less experienced 305
with weightlifting movements. Despite the training difference between the two groups, there 306
were still moderate effect sizes between positions. Similarly, a previously published study 307
evaluated the differences in IMTP and a bent-over deadlift-style “lockout” technique on force 308
production capacity with powerlifters (3). Despite the powerlifters’ lack of experience 309
performing weightlifting movements (i.e. snatch, clean and jerk) and weightlifting derivatives 310
(e.g. mid-thigh pulls), and the large training volumes the lifters had spent practicing deadlift and 311
overloading the deadlift lockout positions, there was still a statistically significant difference 312
(p<0.001) in peak force production between the upright and bent positions and a large effect size 313
(d = 1.23). 314
While the positions used in part 1 of the present study closely mimicked some of the 315
positions used in a study by Comfort et al.(6), force-production differences were observed 316
between the bent and upright positions. Because we did not use specific nuances of methods that 317
were later communicated to us by the authors in part 1, we attempted to replicate exactly the 318
methods used by Comfort et al. (6) in part 2, in order to address the possibility that the method of 319
positioning in part 1 could account for the observed differences. Despite the changes in part 2, 320
and having similar training backgrounds to those of Comfort et al. (6), force production 321
differences remained for the later time points (F200, F250, PF) for all subjects (Hedge’s g of -322
0.54, -0.63 and -0.59, for F200, F250, and PF, respectively). For early time points (F50, F90), for 323
3 of 5 subjects the upright position had substantially greater force values, while the other two 324
subjects there were only small differences favoring the upright position (Hedge’s g of -0.19 and -325
0.35 for F50 and F90, respectively). 326
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
While it is difficult to speculate why no statistical differences in force production 327
between body positions were found in the study by Comfort et al. (6), some possibilities exist. 328
For example, in all of our subjects during the bent position pulls, we observed (from direct 329
observation and video) that nearly all subjects attempted to adjust body position into one 330
resembling the upright position. The increase in joint angles during the pull confirms this 331
observation. In addition, in part 2 of the present study, one of our subjects was unable to 332
maintain the bent position, and immediately shifted during the pull to one that closely resembled 333
the upright position, and was thus excluded from the study. Two more subjects were unable to 334
achieve the correct bent position as specified in the Comfort et al. (6) study, without bending 335
their arms or elevating their shoulder girdle. Had these subjects pulled in the bent position, it 336
seems likely they would have increase their hip angle substantially as their elbows extended and 337
shoulder girdle depressed, ending in a body position similar to that of the other excluded subject. 338
While we are unable to verify if the same body movement issues occurred in the Comfort et 339
al.(6) study, it is at least plausible that some amount of angle change occurred, allowing for the 340
force production between positions to be similar. 341
One particularly interesting finding in the present study is that there is a small amount of 342
extension that occurs at the knee and the hip during the execution of the IMTP (observed with 343
video). While every attempt was made to have the subjects position themselves while using pre-344
tension to minimize slack in the body, the high forces produced during the pull exceed those of 345
pre-tension used to determine position by a large margin. It is possible that these high forces lead 346
to some slight repositioning of the body (e.g. depression of scapula) and change in length of 347
elastic tissues (e.g. decreased height of intervertebral discs, elongation of muscles or ligaments), 348
allowing for some degree of increased knee and hip angle. We attempted to reduce this “slack” 349
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
as much as possible prior to initiation of the pull, but it was apparent that the “pretension” that 350
subjects applied to the bar when setting up their body positions may not have been enough. 351
Subjects did however achieve the desired body position at some point during the pull, whether it 352
was at the start, during the pull, or at the peak extended position. This finding emphasizes the 353
importance of ensuring that very little slack is in the body when determining starting body 354
position and bar height, as well as prior to measured isometric pulls. 355
Some recent research has begun using a “self-selected” position when executing the 356
IMTP (24, 25). One potential issue with using a non-standardized position is that different 357
subjects may be performing worse than would be possible using a standardized and optimal 358
(from a force production perspective) position, especially so if the subject’s chosen position is 359
more bent over than it is upright. The present study indicates that body positioning during the 360
IMTP does matter to force production. Should the “self-selected” position used by any given 361
individual vary between individuals or vary over time in a repeated measures design, it may 362
result in latent variability in performance of which the presence and magnitude is unknown to the 363
researchers. This adds a potentially large source of error into values obtained from the IMTP, 364
thus a self-selected body position is not recommended. 365
366
CONCLUSION 367
The findings of this study indicate that the body position in which the IMTP is executed 368
matters to force production, especially so for subjects with experience with weightlifting 369
derivatives. Furthermore, studies should report both the knee and hip angles used by their 370
subjects for greater ease in comparing results between studies. 371
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
372
PRACTICAL APPLICATIONS 373
In future studies or in practice, we recommend the IMTP be performed with a 120-135° 374
knee angle, and approximately a 140-150° hip angle (upright torso). Bar heights and body 375
positions should be verified under tension, and researchers should expect joint angles to increase 376
slightly during the pull. Consistent bar heights and joint angles should be used when testing over 377
time to ensure that the effect of body position is accounted for. 378
379
REFERENCES 380
381
1. Bakeman R. Recommended effect size statistics for repeated measures designs. Behav 382
Res Methods 37: 379-384, 2005. 383
2. Beckham G, Mizuguchi S, Carter C, Sato K, Ramsey M, Lamont H, Hornsby G, Haff G, 384
and Stone M. Relationships of isometric mid-thigh pull variables to weightlifting 385
performance. J Sports Med Phys Fitness 53: 573-581, 2013. 386
3. Beckham GK, Lamont HS, Sato K, Ramsey MW, Haff GG, and Stone MH. Isometric 387
strength of powerlifters in key positions of the conventional deadlift. J Trainology 1: 32-388
35, 2012. 389
4. Bemben MG, Clasey JL, and Massey BH. The effect of the rate of muscle contraction on 390
the force-time curve parameters of male and female subjects. Res Q Exerc Sport 61: 96-391
99, 1990. 392
5. Cohen J. Statistical power analysis for the behavioral sciences. 1988. 393
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
6. Comfort P, Jones PA, McMahon JJ, and Newton R. Effect of knee and trunk angle on 394
kinetic variables during the isometric midthigh pull: test-retest reliability. Int J Sports 395
Physiol Perform 10: 58-63, 2015. 396
7. Enoka RM. The pull in olympic weightlifting. Med Sci Sports 11: 131-137, 1979. 397
8. Haff GG, Carlock JM, Hartman MJ, Kilgore JL, Kawamori N, Jackson JR, Morris RT, 398
Sands WA, and Stone MH. Force-time curve characteristics of dynamic and isometric 399
muscle actions of elite women olympic weightlifters. J Strength Cond Res 19: 741-748, 400
2005. 401
9. Haff GG, Stone M, OʼBryant HS, Harman E, Dinan C, Johnson R, and Han K-H. Force-402
Time Dependent Characteristics of Dynamic and Isometric Muscle Actions. J Strength 403
Cond Res 11: 269-272, 1997. 404
10. Häkkinen K, Kauhanen H, and Komi PV. Biomechanical changes in the olympic 405
weightlifting technique of the snatch and clean & jerk from submaximal to maximal 406
loads. Scand J Sports Sci 6: 57-66, 1984. 407
11. Hopkins WG. A scale of magnitude for effect statistics. 2014. 408
12. Hornsby WG, Haff GG, Sands WA, Ramsey MW, Beckham GK, and Stone MHE. 409
Alterations in strength characteristics for isometric and dynamic mid-thigh pulls in 410
collegiate throwers across 11 weeks of training. Gazz Med Ital 172: 929-940, 2013. 411
13. Jaric S, Mirkov D, and Markovic G. Normalizing physical performance tests for body 412
size: a proposal for standardization. J Strength Cond Res 19: 467-474, 2005. 413
14. Kauhanen H, Häkkinen K, and Komi PV. A biomechanical analysis of the snatch and 414
clean & jerk techniques of Finnish elite and district level weightlifters. Scand J Sports Sci 415
6: 47-56, 1984. 416
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
15. Kraska JM, Ramsey MW, Haff GG, Fethke N, Sands WA, and Stone MH. Relationship 417
between strength characteristics and unweighted and weighted vertical jump height. Int J 418
Sports Physiol Perform 4: 461-473, 2009. 419
16. McGuigan MR, Newton MJ, and Winchester JB. Use of isometric testing in soccer 420
players. J Aust Strength Cond Assoc 16: 11-14, 2008. 421
17. McGuigan MR and Winchester JB. The relationship between isometric and dynamic 422
strength in college football players. J Sport Sci & Med 7: 101-105, 2008. 423
18. McGuigan MR, Winchester JB, and Erickson T. The importance of isometric maximum 424
strength in college wrestlers. J Sports Sci Med 5: 108-113, 2006. 425
19. McMaster DT, Gill N, Cronin J, and McGuigan M. A brief review of strength and 426
ballistic assessment methodologies in sport. Sports Med 44: 603-623, 2014. 427
20. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, 428
Austria, 2016. 429
21. Stone MH, Sands WA, Pierce KC, Carlock J, Cardinale M, and Newton RU. Relationship 430
of maximum strength to weightlifting performance. Med Sci Sports Exerc 37: 1037-1043, 431
2005. 432
22. Suchomel TJ, Comfort P, and Stone MH. Weightlifting pulling derivatives: rationale for 433
implementation and application. Sports Med 45: 823-839, 2015. 434
23. Tabachnick B and Fidell L. Experimental Design Using Anova. Belmont, CA: 435
Duxbury/Thomson, 2007. 436
24. Thomas C, Comfort P, Chiang C-y, and Jones PA. Relationship between isometric mid-437
thigh pull variables and sprint and change of direction performance in collegiate athletes. 438
J Trainology 4: 6-10, 2015. 439
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
25. Thomas C, Jones PA, and Comfort P. Reliability of the Dynamic Strength Index in 440
collegiate athletes. Int J Sports Physiol Perform 10: 542-545, 2015. 441
442
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Table 1: Reliability results for each subset of analysis Tested Variable Group Position ICC ICC 95% CI CV CV 95% CI Peak Force Both Both 0.991 0.991-0.991 3.4 2.8-4.3 Exp Bent 0.986 0.986-0.987 2.7 1.9-4.6 Upright 0.997 0.996-0.997 1.9 1.3-3.2 Inexp Bent 0.964 0.962-0.965 5.5 3.8-10.3 Upright 0.984 0.984-0.985 2.8 1.9-5.1 Force at 50ms Both Both 0.948 0.947-0.949 8.0 6.6-10.2 Exp Bent 0.803 0.796-0.811 8.9 6.2-15.5 Upright 0.946 0.944-0.948 7.4 5.2-12.8 Inexp Bent 0.903 0.899-0.907 9.6 6.5-18.2 Upright 0.951 0.949-0.953 7.0 4.8-13.2 Force at 90ms Both Both 0.955 0.954-0.956 8.4 6.9-10.7 Exp Bent 0.706 0.694-0.717 12.1 8.4-21.4 Upright 0.979 0.979-0.98 5.7 4-9.9 Inexp Bent 0.939 0.936-0.941 8.9 6-16.8 Upright 0.970 0.968-0.971 6.0 4.1-11.2 Force at 200ms Both Both 0.977 0.977-0.978 5.9 4.9-7.5 Exp Bent 0.943 0.941-0.945 5.2 3.7-9 Upright 0.979 0.978-0.98 5.5 3.8-9.4 Inexp Bent 0.894 0.889-0.899 7.9 5.4-14.9 Upright 0.961 0.959-0.963 4.7 3.2-8.8 Force at 250ms Both Both 0.984 0.984-0.985 4.4 3.6-5.6 Exp Bent 0.977 0.976-0.977 3.3 2.4-5.8 Upright 0.977 0.976-0.977 4.6 3.2-7.9 Inexp Bent 0.946 0.944-0.949 5.2 3.5-9.6 Upright 0.955 0.953-0.957 4.3 2.9-8
Exp: Experienced with weightlifting variations. Inexp: Inexperienced with weightlifting variations. ICC: Intraclass correlation, CV: Coefficient of variation, CI: confidence interval ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Table 2: Force results from isometric mid-thigh pulls in different body positions
PF PFa F50 F90 F200 F250
Bent Exp 3660.7
±612.4
190.8
±31.4
1724.5
±242.5
2058.2
±323.5
2635
±435.3
2722.3
±425.1
Inexp 3108.3
±677.8
174.2
±29.0
1330.8
±251.8
1592.1
±351
2087.9
±360.2
2213.8
±349.9
Upright Exp 4587.1
±981.8
238.9
±49.9
1920.3
±395.5
2441.2
±562
3275.1
±714
3421.7
±688.6
Inexp 3493.9
±568.2
196.5
±22.8
1424.3
±295.1
1755.5
±399.2
2297.4
±388.1
2401.3
±349.1
Exp: Experienced with weightlifting variations. Inexp: Inexperienced with weightlifting
variations. PF: peak force, PFa: allometrically scaled peak force, F50: force at 50ms, F90:
force at 90ms, F200: force at 200ms, F250: force at 250ms.
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Table 3: Results of repeated measures ANOVAs Main Effects Interaction Variable Group Pull Position Group by Pull Position Peak Force F(1,20)=14.9,
p=0.012*, ηg
2=0.25
F(1,20)=45.7, p<0.001*, ηg
2=0.14
F(1,20)=7.8, p=0.01*, ηg
2=0.03 Peak Force (allometrically scaled)
F(1,20)=4.3, p=0.052, ηg
2=0.15
F(1,20)=45.8, p<0.001*, ηg
2=0.18
F(1,20)=6.2, p=0.022*, ηg
2=0.029 Force at 50ms F(1,20)=12.7,
p=0.002*, ηg
2=0.37
F(1,20)=14.2, p=0.001*, ηg
2=0.04
F(1,20)=4.5, p=0.20, ηg
2=0.00 Force at 90ms F(1,20)=11.5,
p=0.002*, ηg
2=0.33
F(1,20)=18.5, p=0.003*, ηg
2=0.07
F(1,20)=3.0, p=0.10, ηg
2=0.01 Force at 200ms F(1,20)=13.7,
p=0.001*, ηg
2=0.37
F(1,20)=41.5, p<0.001*, ηg
2=0.1
F(1,20)=10.7, p=0.004*, ηg
2=0.03 Force at 250ms F(1,20)=14.8,
p=0.001*, ηg
2=0.39
F(1,20)=55.5, p<0.001*, ηg
2=0.12
F(1,20)=18.5, p<0.001*, ηg
2=0.04 * Denotes statistically significant at p<0.05
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Table 4: Joint angle data measured for isometric mid-thigh pulls in each body position
Knee Hip
Start (°) Maximum(°) Change(°) Start (°) Maximum(°) Change(°)
Bent 122.5±7.3 127.3±5.3 5.0±3.3 120.3±6.9 128.7±6.5 7.4±3.8 Upright 120.9±5.2 127.8±5.2 7.0±3.4 138.1±8.9 148.5±6.8 10.4±6.4 Change in angle is the total joint angle change from the starting position of the IMTP (Start) to the highest joint angle observed (Maximum) during the IMTP
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Table 5: Comparison of between force variables for each isometric mid-thigh pull position for each subject
Subject
1 2 3 4 5
PF Bent 3171 4491 2410 3738 5056
Upright 3940 4992 3068 4018 6084 Difference -769 -501 -658 -280 -1028 F50 Bent 1866 2579 1099 1522 1943 Upright 1830 2527 1227 1692 2233 Difference 36 52 -128 -170 -290 F90 Bent 2261 3387 1384 1954 2217 Upright 2275 3308 1729 2124 2951 Difference -14 79 -345 -170 -734 F200 Bent 2682 4016 1956 2619 3320 Upright 2992 4588 2360 2785 4238 Difference -310 -572 -404 -166 -918 F250 Bent 2734 4036 2035 2824 3529 Upright 3222 4831 2480 3019 4328 Difference -488 -795 -445 -195 -799
PF: peak force, PFa: allometrically scaled peak force, F50: force at 50ms, F90: force at 90ms, F200: force at 200ms, F250: force at 250ms.
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Table 6: Individual joint angle data from each subject for each isometric mid-thigh pull position
Knee Hip
Subject Start (°) Max (°) Change (°) Start (°) Max (°) Change (°)
Bent 1 111.5 121.0 9.5 117.0 122.0 5.0 2 125.0 131.0 6.0 126.0 130.0 4.0 3 127.0 127.0 0.0 117.5 123.0 5.5 4 118.0 122.0 4.0 124.0 135.0 11.0
Upright 1 115.0 126.0 11.0 133.0 139.0 6.0 2 122.5 131.0 8.5 143.0 147.0 4.0 3 123.5 126.0 2.5 137.5 147.0 9.5
4 117.5 128.5 11.0 142.5 154.0 11.5 Angle data is missing for subject 5 due to corrupted video file. Change in angle is the total joint angle change from the starting position of the IMTP (Start) to the highest joint angle observed (Maximum) during the IMTP
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Figure 1: Testing Progression of IMTPs
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association
Figure 2: Hedge’s g effect sizes between IMTP body positions with 95% confidence intervals. Exp: experienced with weightlifting group, Inexp: inexperienced with weightlifting group
ACCEPTED
Copyright ª 2017 National Strength and Conditioning Association